Enabling the industrial IoT with cyber security in mind

Industrial Internet of Things (IIoT) networking technology and wireless Machine-to-Machine (M2M) communications solutions are critical to the daily operations of an increasingly connected and industrial world. With a greater dependence on providing reliable and secure high-speed connectivity to personnel, smart devices, machinery and many other geographically dispersed assets, electric utility operators require powerful, yet flexible, communications solutions for the business demands. The overall value proposition that these technologies bring is important for decision-makers to adopt and integrate into their smart grid infrastructures. Some of the main benefits are:

  • Optimising resource utilisation for increased efficiency.
  • Increasing visibility into potential failures and key operational factors through real-time information.
  • Reducing costs and environmental impact through monitoring/control across large geographic regions.
  • Automating safety and security measures to reduce critical event response time.
  • Centralising policy management to create greater adherence and more uniform enforcement of regulatory and compliance factors.
  • Enabling flexibility, scalability and mobility of networks to increase agility to changing market demands

Since electricity transmission and distribution grids are only as efficient, reliable and safe as the equipment and technology used, wireless communications equipment and technologies used in these grids must follow suit. 

Furthermore, as utility operators continue to look for ways to improve their transmission and distribution networks, wireless M2M communications is a critical component because of rapid network installation, lower operational costs, high data throughput, and secure end-to-end transport. In order to achieve these benefits, wireless data networks must be aligned to specific and ranging requirements of the energy grid (generation, transmission, distribution and consumption) in which the wireless communications are utilised and exploited. Wireless IIoT networks transport vast volumes of information from a myriad of applications that are the smart grid; IIoT is the “glue” that holds, binds and delivers the Smart Grid.

In fact, electric utilities are at the forefront of the Industrial Internet of Things with complex and comprehensive wireless networks for advanced metering infrastructure, energy management, distribution and substation automation. It is estimated growth in IIoT applications for utilities and energy industries will increase from 485 million devices in 2013 to over 1.5 billion devices by 2020. This explosive growth in wireless networks, smart sensors and devices, and automated systems requires utilities to address, implement and monitor the security of their data networks because these are the networks providing command and control of critical infrastructure that is the Smart Gird. 

As technology has evolved, so has the intelligence and sophistication of cyber terrorists and their tactics. In 2013, 40 per cent of the cyber-attacks handled by the Department of Homeland Security were against companies in the energy sector; one of the 16 Critical Infrastructure Sectors.

Considering the industrial internet threats to an electric power grid 

According to the Federal Communications Commission (FCC), the two most common threats to wireless networking and communications technologies are Denial of Service (DoS) and Intrusion:

Denial of Service 

Denial of Service is an attempt to make a computer resource or network unavailable to its intended users. DoS attacks could be as simple as jamming an electric or electromagnetic signal or as sophisticated as saturating a system or network with data traffic intended to overwhelm and make the network unavailable. The consequences of DoS attacks range from being simply irritating to destructive. For example, a valve controller does not receive a command to open a value to provide cooling oil resulting in distribution transformer system to overheat resulting in power outages.

Intrusion

Penetrating and intruding into a network or computer resource requires a different level of sophistication. Consequences can range from simply spying or stealing information to corrupting data or maliciously and intentionally causing harm or destruction by taking over network and/or computers and control systems. For example, intentionally penetrating energy management system networks to disrupt and even cut off or redirect the delivery of power to and from substations.

Additional Threats:

  • Sophisticated command and control attacks
  • Packet spoofing
  • Hijacking of sessions / interception
  • Replay attacks
  • Use of worms
  • Trojans and remote controllable Trojans (Back Orifice)
  • Use of a Virus and Anti-forensic techniques
  • Attacks on Domain Name System (DNS) infrastructure

These threats are not limited to wireless communication technologies and need to be addressed as part of a comprehensive cyber security strategy. Organisations must also plan for human error and/or deliberate attacks from internal sources.

Wireless M2M technology and the concept of resiliency  

In order to harden a wireless M2M communications network, and further make IIoT networks and their connected assets less vulnerable to cyber threats, organisations need to implement a multi-layer security strategy, which includes robust networking equipment, hardened communications equipment, network access control and data encryption.

Hardened Communications Equipment

The first line of security is “trusted” networking equipment, where trusted equipment is far more than buying a recognised or favourite brand. Equipment should have the following characteristics.

- Enclosure should have a physical indicator that the mechanical enclosure has not been opened. This can be as simple as a tamper tape or more complex potted solutions.

- The equipment should have a Secure Bootloader, which resides in secure memory so that it cannot be altered or replaced. In addition, the Secure Bootloader validates the executable software image prior to loading it into memory. If the software image is not validated, the device is considered compromised and does not boot or boots with minimum functionality.

- Software images should be encrypted and keyed. This enables the Secure Bootloader to validate the software image prior to execution and prevents the software image from being copied for malicious purposes.

- Equipment must allow data ports to be enabled and disabled so unused ports are unavailable to persons in the proximity of the network equipment.

Consider networking equipment that is Federal Information Processing Standard (FIPS) 140-2 Level 2 validated.

FIPS 140-2 validation consists of four levels of increasing security.

- Level 1 (lowest) – Basic security requirements are specified for cryptographic module.

- Level 2 – Adds requirements for physical evidence of tampering, as well as role-based user authentication.

- Level 3 – Requires physical tampering resistance and stricter identity-based authentication.

- Level 4 (highest) – Adds even more physical security requirements and requires an even greater robustness to the platform, in order to hold up against environmental attacks.

Network Access Control

Network Access Control (NAC) is an essential security feature for wired and wireless networks, since it prevents unauthorised access and intrusion – external or internal – into managed networks. The result is the only right users or devices have access to the right information.

NAC tools are centralised policy and role-based tools, offering combined permissions auditing and management solution for administrators looking to secure their networks against authorised and unauthorised access. As a centralised tool, policy and role base profiles are managed and rapidly deployed across networks of any size.

Policy and role based should include:

- Access restrictions for time of day, day of week and location

- Personnel or device role (e.g. admin, user, guest, process controller) 

- Work Group

- MAC Address of device through which access is initiated

A proven M2M communications network security strategy must go even further and protect data “in transit” as well. Even if an unauthorised device manages to gain access to the M2M communication network, it isn’t necessarily gaining access to the actual data without passing yet another layer of security. 

Data Encryption

Data encryption is an essential part of any security strategy because it does not prevent unauthorised parties from intercepting a message, but encodes a message so only authorised parties can read it, which is a fundamental tenet for wireless networks.

Today, Advanced Encryption Standard (AES) with three keys lengths of 128, 192 and 256 bits is the industry default and used worldwide. As a US Federal Government standard and even used by the National Security Agency, AES can be trusted to protect sensitive information and maintain data privacy.

Wireless links should be encrypted using AES 128 at a minimum.

For a higher level of trust, consider networking equipment that is FIPS 140-2 Level 2 validated.

When a device is FIPS 140-2 validated, a known set of keys and test vectors are passed through the AES algorithm to validate the strength and completeness of the key generation and the encryption algorithm.

Overall, incorporating encryption is a way to ensure data privacy and is yet another way to layer security for information being communicated wirelessly. 

Options for security policies and safeguards

The aforementioned access control and data encryption considerations create resilient and secure wireless networking and communication capabilities for critical infrastructure. However, an IIoT network security strategy also needs to address and implement policies that serve as safeguards, which make it difficult to circumvent security measures and limit the potential impact of a security breach.

Limitation of Permitted Activities

One method to implement safeguards is to limit permitted activities on the wireless network to only those absolutely required to perform or executed the assigned task. This is achieved by disabling ports not required such as TELNET, Port 23.

As an example, a wireless network primarily used for sensor data collection and remote command and control of devices should not allow a hacker that compromised the network to gain access to financial or other critical data. Such a limitation of permitted activities can be achieved through various means of security measures.

Firewalls and Packet Filtering

Packet Filtering and Firewalls based on packet filtering (executed in Layers 4 and lower) are essential features for IIoT devices. IIoT devices typically do not have the processing power for more sophisticated firewall schemes that require packet inspection and are performed at Layer 7.

Virtual Local Area Networks

Virtual Local Area Networks (VLANs) are used to separate the wireless network infrastructure and its management from the production network, devices and/or communication endpoints. By using VLAN’s, we see the introduction of another level of security, especially if combined with Quality of Service (QoS) mechanisms. Think of it as an emergency access to your wireless network infrastructure for remote management and control, in case a Denial of Service (DoS) attack overwhelms the actual payload and production network.  

User Level Access 

By implementing user-level access (password protected), network administrators provide personnel access to wireless network and devices to maintenance personnel, but limited to monitoring system health or performance without opening the system up to misuse because configuration and other privileges are reserved for a different user level and password. 

Access limitation of local ports 

The ability to control who is allowed access from local ports (e.g. through MAC address filtering) or even completely turning off local port access when they are not in use, making it essentially impossible (or at least very hard) for someone who gained physical access to your network infrastructure and devices to get connected and gain access to your private networks.

There are of course several additional safeguarding tactics and strategies worth considering, but these ideas are meant to provide a place from which to expand and adjust as needed.

Finally, in order to complete the M2M security approach, it takes vigilance.  Today’s M2M networks are not heuristic, self-healing, adaptive, self-optimising automatons.  They require an educated observer who is looking for anomalies, aberrations, outliers, exceptions and flat-out failures. M2M requires standards based protocols for determining the health of the network, integrity of the links and performance of the overlying applications.

Going forward with confidence

With the explosive growth of the IIoT market over the next 6 years in sensors, communication equipment (wired and wireless) and the applications to enable ever increasing levels of automation, electric utilities and companies in the energy sector must upgrade existing networks or deploy new networks with security at the forefront. While upgrading or implementing higher levels of security in IIoT and M2M networks, it is a double-edged sword.

- Can’t upgrade devices on a production network because it would disrupt a process

- A cyber-attack is going to disrupt the production network if it occurs.

Today, it is not a matter of “if” a cyber-attack is going to take place, but when.  

The path forward is clear. Develop a cybersecurity plan that secures your expanding IIoT network, but understand threats that exist, and the services and features that are needed from your IIoT devices and networking equipment to lock down communications so they are available to the right persons or devices, when they need to be and can be managed as changes in policies or roles or processes dictate.

Scott Allen is the chief marketing officer at FreeWave Technologies
Image source: Shutterstock/everything possible